price or cost

The neuron selectively activates on numeric tokens—digits and number-related words—wherever counts, prices, or quantities appear in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

mathrm

-1.16

-1.13

 diverse

-1.13

-1.12

-1.10

-1.09

St

-1.06

-1.05

POSITIVE LOGITS

 price

1.91

 prices

1.78

 priced

1.76

 ราคา

1.64

 cost

1.62

 costs

1.60

 costing

1.52

prijs

1.45

售价

1.42

Цена

1.38

Activations Density 0.028%